Interrogates a plan or design through sequential questions until reaching shared understanding.
npx skills add https://github.com/lfuuu/claude-rules --skill grill-me-lfuuu
Please help me install this Agent Skill. Skill: grill-me Source: https://github.com/lfuuu/claude-rules/tree/main/global-skills/grill-me Command: npx skills add https://github.com/lfuuu/claude-rules --skill grill-me-lfuuu
High-intent search queries and answers about installing and using this skill.
Use a structured interrogation approach where each aspect of the plan is questioned one at a time, walking every branch of the decision tree. This Skill automates that process, proposing recommended answers and resolving dependencies between decisions sequentially.
Invoke the grill-me skill and present your plan. The assistant asks one question at a time, offers its recommended answer for each, and continues until every decision branch is resolved and both sides share the same understanding.
Yes. When a question can be answered by investigating the codebase, the skill explores the code itself instead of asking the user. Only questions requiring human judgment or intent are posed directly.
Avoid it for trivial tasks with no branching decisions or when requirements are already fully specified and validated. The interrogation adds the most value for ambiguous, multi-decision plans where hidden assumptions carry implementation risk.
The skill declares support for claude-code and hermes in its frontmatter platforms field. It is distributed as a single SKILL.md file following the agentskills.io standard, so it works in both agents without modification.
SentioLabsCapture approved design approaches in docs/plans and register them with arc plan.
toderianPlan large efforts as maps of decision tickets resolved one at a time.
data-playersAudit code quality across two review phases before merging.
lucas-ferrari-correaGuide structured RFC interviews to generate formal RFC documents.
